Association Between the Metabolome and Cortical Morphology in 5-year-old Children

Abstract

The human brain is continuously developing, from prenatal life to old age, and is subject to regulation by several factors. One such factor that can regulate brain development is metabolism. There is currently very limited knowledge about the influence of the metabolome on cortical morphology - thickness, surface area, volume, and gyrification in 5-year-old children. The aim of this study was to investigate the relationship between the peripheral metabolome and cortical morphology in typically-developing 5-year-old children from the FinnBrain Birth Cohort Study. Previously pre-processed T1-weighted scans of 170 children were used to calculate the local gyrification index (lGI). Multimodal neuroimaging analysis was performed to obtain 10 multimodal components combining 4 cortical modalities: cortical surface area, thickness, volume, and lGI. Associations were examined both with metabolite clusters formed by Gaussian mixture modelling and with individual metabolites using partial correlation analysis adjusted for age, sex, and BMI, and elastic net regression. Candidate clusters and metabolites were then chosen for vertex-wise analysis (VWA) to determine whether their associations were localised to a particular cortical tissue property in any particular part of the brain. VWA identified a significant association between cluster 2, containing lysophospholipids, and surface area in the left cuneus, superior parietal, and lateral occipital regions. Metabolite unknown_21538 was identified to have a significant association with the surface area in the left parietal cortex. All analyses were adjusted for age, sex, and BMI. Lysophospholipids play an important role in cortical development by surface remodelling, myelination, and synaptic refinement. Identification of the unknown metabolite may highlight an important biological association in cortical development.